user-research-prior-art-survey

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (medium risk: 0.30). The workflow ingests outsider-authored full text from external sources it fetches—per-angle selection (e.g., references/angles/a1.md and references/angles/a2.md) can resolve and then deep-read a candidate’s full text, with the admission rule explicitly requiring method-bearing full text retrieval.